Study On Adaptive Anti-Jamming And Blind Source Separation Application Technology

The spatial anti-jamming technology has been a hot research topic and a key point in radar engineering applications recently.Effective suppression of external active interference is a critical evaluation index of radar working capability.The most commonly used sidelobe interference suppression method in engineering applications is adaptive sidelobe cancellation.This method is used as a spatial adaptive dimensionality reduction process.The selection of auxiliary channels directly affects the interference suppression effect.Therefore,it is very vital to study different forms of auxiliary channels and its influencing factors such as number,layout,etc.on the performance of interference suppression to guide project realization.Also,because the traditional spatial filter method is incapable of main-lobe interference,the study of anti-main-lobe interference methods is of great significance to improve radar performance in complex electromagnetic interference environment.The work of this thesis is mainly developed from the following two aspects:1.For sidelobe interference,the traditional adaptive beamforming technology can generate nulls in the interference direction of the equivalent receiving pattern and do not affect the radar detection capability for targets in the main-lobe.This thesis takes the sidelobe cancellation as an example to study specific methods of dimensionality reduction from the array-element-domain and the beam-domain,and analyzes their anti-jamming capabilities.For array-element-domain dimensionality reduction processing methods,the influence of the number and position of the auxiliary array elements on the interference suppression performance is discussed,and a better method of auxiliary array element selection is given based on the simulation results.For beam-domain dimensionality reduction processing methods,the interference suppression performance of the Adams method and Gabriel method are discussed.Simulation results show that the Adams method can effectively suppress dense interference with very few auxiliary beams,but for sparse interference,sidelobe uplift and main-lobe widening need to be considered comprehensively to obtain a better adaptive pattern.The Gabriel method can obtain better interference suppression effects when the estimated interference angle error does not exceed half the beamwidth,but the number of array elements required to synthesize the auxiliary beam must be guaranteed not less than the number of interferences.2.When there is interference from the main lobe,the traditional adaptive beamforming method will cause the sidelobe to rise sharply,and the directional pattern will be distorted,which will seriously affect the detection ability of the target.Blind Source Separation(BSS)can effectively separate the signal sources in different directions as long as the signal sources are independent of each other when the manifold information of array is lacking.Because the BSS method has a certain angle super-resolution capability,this method provides a new idea for suppressing radar main-lobe interference.In this thesis,the BSS-based Second Order Statistics(SOS)algorithm is introduced into the main-lobe interference suppression.First,the influence of factors such as the signal-to-noise ratio,the angle between target and interference,and the beam direction on the separation effect is simulated and analyzed.Then,because of the large difference between the target signal separated by the SOS algorithm and the original signal in amplitude,an amplitude compensation method is proposed.This method uses the constant obtained by normalization to compensate the separated signal,which overcomes the problem of inaccurate estimation of the signal source amplitude in the conventional BSS method.In addition,this thesis proposes an anti-main lobe interference method based on BSS and spatial matched filtering.This method uses the BSS method to separate the interference in the main lobe from the target signal,and then the equivalent array receiving signal is constructed for spatial matched filtering,so as to obtain an accurate estimate of the target angle.The simulation analysis shows that the two proposed methods can better solve the uncertainty of the separated signal amplitude in the traditional BSS method.Finally,an anti-main-lobe interference method based on the super-resolution algorithm is proposed.In this thesis,the Multiple Signal Classification(MUSIC)algorithm is used to estimate the mixing matrix,which can be used to separate the target and the interference signal in the main lobe.This method has a high estimation accuracy of target signal amplitude and is suitable for the array antenna’s high calibration accuracy.Compared with the traditional BSS method,this method is sensitive to amplitude and phase errors,and the angular resolution is slightly worse.